Adjuvant pembrolizumab for participants with hepatocellular carcinoma and complete radiologic response after surgical resection or local ablation: The phase 3 keynote-937 study.

Eligible pts were aged ≥ 18 years with confirmed HCC, complete response after surgical resection or local ablation, ECOG PS 0 - 1, and Child-Pugh liver class A. Pts with past or ongoing hepatitis C or controlled hepatitis B virus infection could enroll after meeting prespecified criteria. All pts were randomized 1:1 to pembrolizumab 200 mg or placebo IV Q3W until disease recurrence, unacceptable toxicity, intercurrent illness, withdrawal, or up to 17 cycles of pembrolizumab or placebo. Randomization was stratified by geographic region, prior local therapy (resection vs ablation), recurrence risk, and alpha-fetoprotein level at diagnosis. The primary endpoints were recurrence-free survival (RFS) by imaging (BICR) or pathology, and overall survival (OS). Safety was a secondary endpoint. Key exploratory endpoints included distant metastases-free survival (DMFS) and time to recurrence. The data cut-off date was Mar 20, 2025.

A total of 959 pts were randomized (476 pembrolizumab; 483 placebo). At IA3, median follow-up was 50.7 months (range, 31.6 – 69.1). Median RFS was not statistically different with adjuvant pembrolizumab vs placebo (46.7 mo vs 45.5 mo; HR 1.06; 95% CI, 0.88-1.26; P=0.719 [P-value boundary for significance of 0.0187]), with 48 mo RFS rate of 50% vs 50%. As the RFS hypothesis was not met, OS was not tested per multiplicity and the study will not proceed to final analysis. Median OS was not reached (NR) in either arm (HR 1.08; 95% CI, 0.81-1.43; P=0.704 [nominal]), with 48 mo OS rate of 79% vs 81%. Median DMFS was NR (95% CI, 58.7 to NR) with pembrolizumab vs NR (95% CI, 59.0 to NR) with placebo (HR 0.98; 95% CI, 0.77-1.24), with 48 mo DMFS rate of 71% vs 70%. Median time to any recurrence was 52.5 mo vs 50.1 mo. Grade ≥3 adverse events (AEs) occurred in 32% vs 22% pts, respectively, with grade ≥3 drug-related AEs occurring in 14% vs 5% of pts. There were no drug-related deaths.